Mandatory governance reform and corporate risk management
Introduction
Do self-interested managers hedge too much or too little relative to the best interest of shareholders? Corporate risk-taking and risk management are a central issue in modern finance, but the fundamental relationship between agency problems and corporate risk policies is not yet well understood. Notably, theory and empirical evidence suggest that this relationship is ambiguous, expressed in two popular but conflicting views. In one view, managers hedge too little due to agency problems such as career concerns leading to short-termism and convex incentives; in this view, firms reduce financial corporate risk when exposed to more stringent governance requirements that safeguard shareholders' interest (Froot et al., 1993; Myers, 1977; Leland, 1998). In the contrasting view, agency models predict the opposite: when managers are imperfectly diversified or overly risk-averse, they may choose hedging policies that suit their own interests and ‘overhedge’ in the absence of safeguards for shareholders' interests. Stronger governance will then lead to less hedging (Stulz, 1984; Smith and Stulz, 1985). There is substantial empirical evidence in support of either view.
Not only is there a strong and unresolved contrast between these views, but knowledge on this issue also suffers from methodological shortcomings: earlier work does not establish causality of the relationship between governance and risk management based on modern methods of identification. Our paper attempts to make headway on this issue by studying a large-scale natural experiment, the adoption of the Sarbanes-Oxley (SOX) Act of 2002 in the U.S. and its sweeping governance changes. No earlier empirical study investigating the impact of governance on hedging uses a quasi-experimental setting.
We find strong evidence that improvements in governance lead to less foreign exchange risk and more foreign exchange derivatives hedging. This finding is robust whether we look at the initial reform shock or the heterogeneous and staggered adoption of the induced governance reforms, whether we focus on exogenously imposed governance reforms, isolate board and compensation as the major transmission channels of the reform, or add voluntary governance improvements.
A major contribution of our paper is that we are the first, to the best of our knowledge, to address concerns about endogeneity in the governance-risk management relationship by using a rigorous identification strategy based on an exogenous governance shock. This approach fills an important gap, in light of longstanding concerns that corporate governance is highly endogenous (see e.g. Himmelberg et al. (1999) for an early critique of neglected endogeneity in governance research). We do so by undertaking a systematic difference-in-difference panel analysis around the SOX quasi-experiment and studying its impact on corporate hedging, exploiting the fact that it affected firms differently according to their pre-reform governance practices. In addition, we make two other novel contributions to the literature. First, we argue that firms with a larger ex-ante distortion in hedging should respond more strongly to an exogenous governance reform, which allows us to refine our hypotheses by incorporating pre-reform firm heterogeneity. We measure such heterogeneity across firms along two dimensions, their exposure to foreign markets and their CEO incentives and CEO characteristics. For both dimensions, we find evidence consistent with our predictions, and hence are able to confirm our main finding with rich cross-sectional evidence. Second, we distinguish between financial and operational hedging strategies, and argue that their deployment should follow different trajectories, with financial hedges being deployed more quickly in the post-reform adjustment and assuming the bulk of the early post-treatment adjustment. Our panel data allow us to study different adjustment speeds, and we present evidence consistent with this prediction. We also present a number of ancillary results, including the finding that the level of adoption of required governance reforms is a stronger predictor of hedging than voluntary governance improvements.
In more detail, we deploy a difference-in-difference approach with firm fixed effects to study the quasi-experiment of the 2002 Sarbanes-Oxley Act, arguably the most sweeping reform of governance mandates in a developed market in the past 30 years. We analyze a broad balanced panel of U.S. corporations drawn from the Russell 1000 index, and differentiate our measures of treatment, the reform impact on governance, in two ways. First, we argue that when firms are assigned to different groups according to their treatment status, it is essential to distinguish between the (regulator's) intention to treat, and firms' compliance with the treatment, in the parlance of the literature on heterogeneous treatment effects (e.g. Angrist and Pischke, 2009), since not all firms adopt the new rules fully and immediately. To account for the intention to treat, we consider the pre-reform governance gap that focuses on the exogenous nature of the quasi-experiment treatment shock by comparing the difference between post- and pre-reform hedging of firms with high treatment intensity (large gap) to the difference of firms with little or no treatment (small gap). We measure compliance with the treatment with the actual reform implementation that focuses on the effective year-by-year policy adoption of the governance measures, and thus accounts for the substantial heterogeneity in compliance across firms.
Second, we address the observation that the Sarbanes-Oxley Act was effective because it triggered broad changes, with firms adopting concomitantly many measures beyond the changes imposed by the reform, under pressure from shareholders and the public. There is a trade-off between a narrow measure that contains only mandated changes and a broader measure that better captures the full scope and heterogeneous adoption of the reform. We resolve this trade-off by using a broad array of five governance indexes, varying along a continuum between exogeneity and representativeness. Our most restrictive index includes only the ten governance attributes that the SOX reform rendered obligatory by 2004 (REG10), with a strong focus on board policies. Our second index (MANDATE-INDUCED) adds ten governance attributes that were so tightly connected with the SOX law that their simultaneous adoption was highly likely. In an alternative approach, we use two indexes limited to the two main channels of reform, the Board of Directors and executive compensation (BOARD&COMP15 contains all mandated/mandate-induced board and compensation measures; BOARD&COMP22 adds voluntary measures). Finally, we use a comprehensive index of 41 governance measures, GOV41.
We refine our main hypotheses by considering the role of cross-sectional variation in pre-reform distortions in risk management. We argue that, the larger the level of distortions prior to the reform, the larger should be the adjustment in hedging triggered by the governance reform. We are interested in exploring how governance changes affects financial and operational hedging, and whether financial hedges are adjusted faster than operational hedges. To address this issue, we deploy a dual measurement of corporate hedging, the use of FX derivatives as a measure of financial hedges, and the exposure to foreign exchange rate movements as a comprehensive measure. Foreign exchange exposure is appealing and widely used as it reflects all of the different instruments of corporate risk management, including financial and operational hedges. The use of derivatives on the other hand isolates intentional risk management and financial hedges.
Our findings are as follows. We find strong evidence that a weak governance environment is associated with insufficient attention to risk management – or ‘underhedging’ by self-interested managers. The use of foreign exchange derivatives is intensified and foreign exchange exposure is reduced after the reform. The results are the same for initial governance gap or for actual implementation. The effects are highly consistent and significant in 19 out of 20 multivariate tests (for 2 hedging variables, 5 indexes and 2 measures of governance change). Our results are driven by mandated and mandate-induced attributes, as voluntary changes are less significant.
The economic magnitude of the effect of governance changes on risk management in our study is large: according to our estimates, the median increase in REG10 attributes between 2002 and 2007 leads to a 20.4% decrease in foreign exchange exposure and a 52.2% increase in Derivatives Mentions, and the median increase in MANDATE-INDUCED (GOV41) to a 21.0% (20.4%) decrease in FX exposure and a 31.1% (48.4%) increase in Derivatives Mentions.1
The next important step is to investigate whether our baseline results are confirmed when we explore the governance-risk management relationship exploiting firm-level heterogeneity. We find that firms with large exposure to foreign markets, i.e. firms that trade globally or have a large foreign input exposure, exhibit a strong reaction of hedging to governance reform and whereas firms with little exposure react significantly less. When we look at differences in managerial incentives, we find that firms where CEOs hold large stock option positions react more strongly to governance reform, and a weaker effect in the opposite direction when CEOs hold large equity stakes, consistent with theoretical predictions on executive compensation. When we deploy a more granular time-trend analysis, we find a noticeable difference in the speed of adjustment between the use of derivatives and FX exposure: whereas governance reform exerts an immediate effect on the use of FX derivatives, consistent with the notion that financial hedges are adjusted quickly, we show that foreign exchange exposure, which comprises operational hedges, is adjusted much more slowly. Also, when sorting firms according to major characteristics, such as governance, size, value/growth, we find that large firms show a stronger hedging reaction to the governance shock than smaller firms, but find little differences for the growth/value dimension or the G-index.
The remainder of the paper is structured as follows. The next section discusses the institutional background, prior literature and hypotheses. Section 3 describes our data and methodology. Section 4 presents summary statistics and univariate findings, and Section 5 discusses the multivariate analysis of our two identification approaches. Section 6 investigates the role of firm heterogeneity. Section 7 explores the different adjustment speeds of financial and operational hedges. Section 8 covers robustness issues, and Section 9 concludes.
Section snippets
The Sarbanes-Oxley Act and corporate governance reform
The Sarbanes-Oxley Act (SOX) of 2002 in the U.S. was an unprecedented overhaul of corporate governance standards in reaction to a wave of fraud and accounting scandals in the early 2000s, the most emblematic ones engulfing Enron and WorldCom. The legislation included a number of stringent provisions on board composition, board oversight and compensation, intended to safeguard investors and to mitigate governance problems. The main thrust of the Sarbanes-Oxley mandates and associated revisions
Data and measures of corporate hedging
We start our sample construction from the set of all Russell 1000 firms that were listed for the period 2000–2007. We remove financial firms and REITs (which have unique governance structures), leaving us with 786 remaining firms.4 Financial information – including stock price and financial statement data – is from Compustat and Datastream, and the governance data are drawn from RiskMetrics
Summary statistics and univariate analysis
Table 1 presents in Panel A the mean, median and standard deviation by year (2002–2007) for our five indexes and two complements, ordered from the narrowest, purely mandatory REG10 index to the comprehensive GOV41 index. There was a significant rise over time in all indexes as firms worked towards compliance, as the t-tests (difference in means) and Wilcoxon rank sum test statistics (medians) in column (8) show for the 2007 vs. 2002 difference. REG10 shows the strongest increase in means, an
Analysis of the initial governance reform gap
A graphical analysis of the parallel trends assumption confirms that our two measures of risk management do not exhibit a pre-trend convergence prior to the SOX reform. Fig. 1 shows the graphs for both hedging measures, for the most narrow governance index REG10 (we plot the top tercile for any of the governance measures against the values of the two remaining terciles). No apparent converging (or diverging) trend is discernible in the three years prior to the SOX reform. In contrast, a trend
Heterogeneous firm exposure to foreign markets
In this Section, we explore whether the results of our main regressions hold up when we test them against cross-sectional evidence. Firms exhibit different levels of pre-reform exposure to foreign exchange risk and to distortions in managerial incentives, and hence their reaction to a uniform governance shock should not be the same. This offers opportunity to develop additional tests, and to better understand the mechanics and severity in distortions to optimal hedging policies. As we express
Financial and operational hedges and differences in adjustment speed
Hypothesis 3 postulates that firms will react more rapidly to a governance reform shock by adjusting their financial hedges compared with their operational hedges. To investigate the speed of reaction and understand the trajectory of firm's reaction to the SOX reform, we use a dynamic difference-and-difference approach and interact our governance indexes with year dummies 2002 to 2006 in Table 10. The first term, the governance index without time interaction term, indicates the time-invariant
Firm fixed effect and industry-year fixed effect panel estimations
We double check the robustness of our results by reestimating our main regressions of Table 3, Table 4 in panel estimations with firm fixed effects in lieu of deviations-from-means. The results are presented in the Internet Appendix (Tables IA.9 and IA.10) and show that our findings are indeed robust under this alternative specification.21
Conclusion
Following the corporate governance scandals of the early 2000s, the strong regulatory response in the form of the 2002 Sarbanes-Oxley Act led to a comprehensive reform process that strengthened board supervision and limited managerial discretion. We use this large exogenous shock as a unique, yet unexplored opportunity to advance our understanding of the relationship between corporate governance quality and foreign exchange risk management. Using a large panel of US firms, our strategy is to
Acknowledgement
Hege acknowledges funding from the European Research Council, ERC FP7 grant No. 312503-SolSys, and from the ANR, grant no. ANR-17-EURE-0010 (Investissements d'Avenir).
References (76)
- et al.
Does governance travel around the world? Evidence from institutional investors
J. Financ. Econ.
(2011) - et al.
Do corporate governance mandates impact long-term firm value and governance culture?
J. Corp. Finan.
(2019) - et al.
Exchange rate exposure, hedging, and the use of foreign currency derivatives
J. Int. Money Financ.
(2001) Corporate hedging and speculation with derivatives
J. Corp. Finan.
(2019)- et al.
Methods for multicountry studies of corporate governance: evidence from the BRIKT countries
J. Econ.
(2014) - et al.
Do takeover laws matter? Evidence from five decades of hostile takeovers
J. Financ. Econ.
(2017) - et al.
Exchange rate exposure
J. Int. Econ.
(2006) - et al.
Do managerial risk-taking incentives influence firms’ exchange rate exposure?
J. Corp. Finan.
(2017) - et al.
Playing it safe? Managerial preferences, risk, and agency conflicts
J. Financ. Econ.
(2016) - et al.
CEO compensation and corporate risk: evidence from a natural experiment
J. Account. Econ.
(2013)
Understanding the determinants of managerial ownership and the link between ownership and performance
J. Financ. Econ.
Offshore activities and financial vs operational hedging
J. Financ. Econ.
Foreign exchange exposure and multinationality
J. Bank. Financ.
Theory of the firm: managerial behavior, agency costs and ownership structure
J. Financ. Econ.
Currency hedging and corporate governance: a cross country analysis
J. Corp. Finan.
An integrated model of multinational flexibility and financial hedging
J. Int. Econ.
Foreign exchange risk exposure: survey and suggestions
J. Multinatl. Financ. Manag.
The determinants of corporate borrowing
J. Financ. Econ.
Exposure to currency risk: definition and measurement
Financ. Manag.
The use of foreign currency derivatives and firm market value
Rev. Financ. Stud.
The use of foreign currency derivatives, corporate governance and firm value around the world
J. Int. Econ.
Valuation of the operating flexibility of multinational corporations
J. Int. Bus. Stud.
Mostly Harmless Econometrics, an Empiricist’s Companion
Corporate hedging and shareholder value
J. Financ. Res.
Shock-based causal inference in corporate finance and accounting research
Crit. Financ. Rev.
The exchange rate exposure puzzle
Manag. Financ.
The effects of derivatives on firm risk and value
J. Financ. Quant. Anal.
Managerial entrenchment and capital structure decisions
J. Financ.
Enjoying the quiet life? Corporate governance and managerial preferences
J. Polit. Econ.
Exchange rate exposure: A simple model
Estimating exchange rate exposures: issues in model structure
Financ. Manag.
Wharton survey of derivative usage by U.S. non-financial firms
Financ. Manag.
A view inside corporate risk management
Management Science
Wharton 1995 survey of derivative usage by US non-financial firms
Financ. Manag.
Wharton survey of financial risk management by US non-financial firms
Financ. Manag.
Pass-through and exposure
J. Financ.
Board composition and corporate use of interest rate derivatives
Journal of Financial Research
Why do firms hedge? An asymmetric information model
J. Fix. Income
Cited by (5)
Shareholders' political hierarchy and regulatory enforcement: Evidence from corporate risk management
2024, British Accounting ReviewEconomic policy uncertainty and corporate risk-taking: Loss aversion or opportunity expectations
2021, Pacific Basin Finance JournalCitation Excerpt :With regard to executive characteristics, executive power (i.e., the social capital of the Chief Executive Officer [CEO]) (Ferris et al., 2017), gender (Faccio et al., 2016), overconfidence (Malmendier and Tate, 2005, 2008; Malmendier et al., 2011) and other characteristics will impact corporate risk-taking. Some scholars have conducted relevant research regarding the enterprises' external environment, including government intervention (Francis et al., 2009), marketization degree (Zhang et al., 2021), rule of law (Hege et al., 2021; Koirala et al., 2020), social culture (Díez-Esteban et al., 2018; Tran, 2019), political connections (Khaw et al., 2016; Shi et al., 2019), and the nature of property rights (Boubakri et al., 2013; Zhang et al., 2021). Studies signify that rising economic policy uncertainty (EPU) will raise the cost of corporate debt while simultaneously reducing corporate debt ratings.
Risk management
2024, Handbook of Corporate FinanceHow did the Sarbanes–Oxley Act affect managerial incentives? Evidence from corporate acquisitions
2022, Review of Quantitative Finance and Accounting